Artificial-Intelligence (ai) System for Generating Event Analytics

This application claims priority to U.S. Provisional Patent Application Ser. No. 63/678,494 filed Aug. 1, 2024.

This application relates to: U.S. patent application Ser. No. 19/209,013 filed May 15, 2025, which is a continuation of U.S. Pat. No. 12,322,216 filed Oct. 24, 2022, which is a continuation of U.S. Pat. No. 11,501,582 filed Nov. 30, 2020, which claims benefit of U.S. Provisional Application 62/942,156 filed Dec. 1, 2019; U.S. patent application Ser. No. 17/841,507 filed Jun. 15, 2022, which claims benefit of U.S. Provisional Application Ser. Nos. 63/210,922 filed Jun. 15, 2021 and 63/299,340 filed Jan. 13, 2022; U.S. patent application Ser. No. 19/016,617 filed Jan. 10, 2025, which claims benefit of U.S. 63/620,157 filed Jan. 11, 2024 and which is a CIP of U.S. patent application Ser. No. 17/971,983 filed Oct. 24, 2022, which is a CIP of U.S. Pat. No. 11,501,582 filed Nov. 30, 2020, which claims benefit of U.S. Patent Application Ser. No. 62/942,156 filed Dec. 1, 2019, and which is a CIP of U.S. patent application Ser. No. 17/841,507 filed Jun. 15, 2022, which claims benefit of U.S. Provisional Application Ser. Nos. 63/210,922 filed Jun. 15, 2021 and 63/299,340 filed Jan. 13, 2022, and which claims benefit of U.S. Provisional Application Ser. No. 63/678,494 filed Aug. 1, 2024.

This application hereby incorporates by reference the above-listed applications in their entireties as if fully set forth herein.

This disclosure is protected under United States and/or International Copyright Laws. @ 2022*. All Rights Reserved. A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and/or Trademark Office patent file or records, but otherwise reserves all copyrights whatsoever.

In an embodiment, a system uses Bluetooth® and/or Wi-Fi® technology to identify participants such as athletes competing in, or otherwise participating in, an event or a contest such as an athletic event or athletic contest. Components of the system include beacons, software applications residing in the cloud, and gateways. Beacons transmit their identifications (e.g., the IDs of the athletes carrying the beacons) in the Bluetooth® band, and each gateway picks up the IDs of one or more beacons in close proximity to the gateway. Beacons are attached to the participants (e.g., pinned onto their shirts, strapped to their wrists) to identify the participants individually, and gateways are distributed along the competition course (e.g., along a running-race course) to pick up the signals from the beacons. Where the event is a running race, an operator (e.g., a race organizer or sponsor) specifies (1) the course map, and (2) the location of the dropped gateways on the course. While the race is ongoing, the gateways, the positions of which are fixed, track the position of each runner over time by tracking the signal emitted by the runner's beacon. From the position-and-corresponding-time data for a runner, a system, such as a cloud computer system, can calculate metrics (e.g., analytics) such as the runners average speed or an average pace at any particular stretch along the course. With additional data (e.g., a runner's height, weight, the wind direction and velocity), a system, such as a cloud computer system, can calculate additional metrics (e.g., performance vs. body weight, performance vs. outside temperature), for example, by leveraging AI.

In an embodiment, a system is configured to use humanoid (e.g., facial) recognition, optical-character recognition (OCR), or both humanoid recognition and OCR instead of, or in addition to, Bluetooth® technology to identify athletes competing in an event or a contest. Other than being so configured, the system can be similar to the embodiment of the system described in the preceding paragraph. Or the system can omit the beacons and at least some of the gateways to reduce a cost of the system or a cost of implementing the system.

In an embodiment, a system includes an image-capture device and a computing circuit. The image-capture device is configured to capture an image of a participant identifier as a participant carrying the participant identifier crosses a finish line of an event course. And the computing circuit is configured to determine, in response to the captured image of the participant identifier, an identity of the participant, and to store, in a memory, the determined identity of the participant and a time at which the participant crossed the finish line.

This application is intended to describe one or more embodiments of the present invention. It is to be understood that the use of absolute terms, such as “must,” “will,” and the like, as well as specific quantities, is to be construed as being applicable to one or more of such embodiments, but not necessarily to all such embodiments. As such, embodiments of the invention may omit, or include a modification of, one or more features or functionalities described in the context of such absolute terms. In addition, the headings in this application are for reference purposes only and shall not in any way affect the meaning or interpretation of the present invention.

In an embodiment, a system is configured to collect data regarding an event (e.g., an athletic event such as a running race) and to generate, for event participants (e.g., runners) and from the collected data, analytics regarding each participant relative to the event, and regarding the event itself. The system includes first devices (typically wireless devices), called beacons, associated with the event participants for identifying, electronically, the participants as they participate in the event, second devices (typically wireless devices), called gateways, distributed around the geographic region where the event takes place and configured to receive raw data from the beacons and to process at least some of the raw data, and one or more computing machines (e.g., cloud-based) configured to receive at least some of the raw data and the processed data from the gateways, to analyze the raw and processed data, to maintain a database of a number of similar events and participants of those events, and to generate analytics in response to the analyzed data and data from the database. The one or more computing machines may perform one or more of these operations by executing, or otherwise using, artificial intelligence (AI).

For example, a 5K race includes five hundred participants. Each participant wears, or otherwise carries, a Bluetooth® compatible beacon that transmits information such as a unique identity (ID) of the participant. The beacon also can transmit other information including participant location (e.g., if the beacon includes a GPS locator), air temperature or humidity in the vicinity of the beacon, or a number of steps taken by the participant over one or more periods of time. Due to popularity of RFID readers and scanners used in participatory athletic events, a Bluetooth® beacon could be armed or loaded with an RFID tag, which could make the beacon readable by both Bluetooth® and RFID scanners/gateways. That is, in an embodiment, a beacon can be capable of identifying an event participant using Bluetooth® signals, one or more passively read RFID tags, or both Bluetooth® signals and one or more passively read RFID tags. Furthermore, such a beacon can be mounted, or otherwise attached, to a slap band or slap bracelet (also called a snap bracelet) to facilitate wearing of, and removal of, the beacon by an event participant. Alternatively, instead of, or in addition to, Bluetooth® beacons, one can use OCR to read a participant's identification number and other data from a bib worn by the participant.

The gateways, which are distributed along the racecourse (e.g., one gateway every one hundred meters along the racecourse), are configured to receive the signals (e.g., Bluetooth® signals) from the beacons as the beacons pass by the respective gateways. For example, a gateway may be able to recover data from a signal received from a beacon while the beacon is within, e.g., thirty meters of the gateway.

In an embodiment, each gateway can perform some analytics, like determining an average speed or average velocity of a participant as he passes the gateway. And where the gateways can communicate with one another or with a common computing circuit, each gateway (alternatively the common computing circuit), can perform other analytics like determining a participant's leg/split times.

In an embodiment, a gateway can be an electronic circuit that includes a processing circuit (e.g., a microprocessor or microcontroller) and a Bluetooth® transceiver (and/or possibly an RFID tag reader or Wi-Fi® transceiver). For example, a gateway can include a Raspberry Pi® microcontroller that is programmable, or that otherwise is configurable, to perform a variety of functions or operations including one or more of the functions or operations described herein.

In a similar embodiment, microcontroller-based gateways are deployed on a participatory athletic event course to capture data that can determine the crossing times of the participants at points of interest (e.g., checkpoints distributed uniformly or nonuniformly along the event course). The data can be generated by beacons that communicate with the gateways using Wi-Fi®, Bluetooth® (e.g., low-energy Bluetooth® (BLE)), RFID, humanoid recognition, OCR, or by any other devices or systems operating on public or commercial bands and that can be used as beacons. For example, a system can use the Bluetooth® band of communication by using Bluetooth® gateways and Bluetooth® beacons to individually identify and determine the crossing time of each participant at the aforementioned checkpoints. The system can use the data captured from these intermediary checkpoints on a race course to create a new generation of race score cards for each participant and to perform analytics using an AI engine that allows the AI system to access, or otherwise to utilize, a large proprietary knowledge base of performances by event participants (e.g., race runners) from a wide range of backgrounds, builds, makeups, age, and conditioning at participatory endurance events to provide customized analytics to each individual user (e.g., event participant) of the AI system.

2 FIG. Bluetooth® gateways can be used to capture Bluetooth® signals from any device with a Bluetooth® transmitter, which device can be configured as a beacon. In an embodiment, the Bluetooth® gateways can calculate the crossing times of the participants, communicate and transmit this data to the cloud, and store the data on the local gateway so as to preserve the information in case of loss of connectivity. In an embodiment, a microprocessor or microcontroller is used to supplement or to replace the gateway (e.g., the gateway can include the microprocessor or microcontroller). And a small computer, such as a Raspberry Pi®, an Nvidia computer integrated circuit (IC), or a similar device can be used as a microprocessor or microcontroller. A Raspberry Pi®, acting as a gateway, can store almost unlimited (e.g., limited by available volatile memory on the Raspberry Pi®) Bluetooth® data (e.g., data from one or more of the Bluetooth® beacons) onboard, transfer the data to the cloud, and perform functions to sort and process the data with local and customized applications onboard. The data captured and processed by the microprocessor or microcontroller can be used to determine the crossing times of a race participant at checkpoints set along the course. Use of microprocessors or microcontrollers can allow event (e.g., race) organizers to capture data and perform calculations on the data to create a new generation of score cards. An example of such a score card is shown in.

In an embodiment, a computer-based AI system (e.g., an AI engine) can generate a new generation of score cards for the participants of endurance events (e.g., running races), where data from microcontroller-(or microprocessor)-based gateways as described herein can provide a more-detailed breakdown of a participant's performance at an endurance event. By placement of the microcontroller-(or microprocessor)-based gateways along the course (e.g., every ten meters, twenty meters, one hundred meters, kilometer, or mile, or at bottom and/or top of each climb or descent), the AI system can deliver a more-descriptive and -interesting set of data to each participant. A score card can be broken up into a number of legs of the event course, where each leg can have a set of relevant and descriptive indicators such as average slope, elevation change, temperature, wind, or wind direction.

In an embodiment, an AI engine can access a large knowledge base of the performances of participants at the endurance events and can present a participant with an AI command-line interface or prompt to allow the participant to request multivariant analytics, data analysis, basic reporting, or other information. The AI system can include a command-line prompt that refers to an AI-prompt interface that utilizes a knowledge base to perform the tasks and functions requested by the user (e.g., an event participant such as a runner). The knowledge base utilized by the AI system is made up of data from gateways deployed on the course, and a large historical data set of runners from a variety of demographics who opted in to join the knowledge base with one or more of the following: gender, age, height, weight, and/or other physiological or telemetry data such as heart rate (e.g., resting heart rate, just-after-a-race-has-finished heart rate). After each event, the AI system can be used to provide basic statistical reports on finishers, pace, distributions, and variances from norm. One (e.g., a participant in the event) also can prompt the AI system to compare the performance of an individual participant and examine and predict variations of his/her performance by correlating the participant's (e.g., runner's) data to the data stored in a large knowledge base that has been created over a long period of time. The large knowledge base can have data from participants of all ages, backgrounds, body composition, heights, genders, etc. The AI system also can examine, for example, the impact of more or less heat on a particular runner, the impact of more or less body weight on the pace of a particular runner during a climb, or the impact of age on the performance of a particular runner.

In an embodiment, where the event is a running race, the gateways for the Bluetooth® signal detection are relatively inexpensive; therefore, distributing several of them throughout the racing course enables the capture of data from key spots along the course and the creation of reports (e.g., score cards) for runners. If there is a race with a straightaway and a climb, followed by another straightway, one could “drop” a couple of gateways along the straightaways, and have at least a few gateways along the climb to time runners at or near the bottom, at or near the midpoint, and at or near the top of the climb. After the race, a runner's score card can reflect his/her splits for the climb, for the straightaways, and for any race segments or legs with points (e.g., locations) from which the system (e.g., a Bluetooth® system and/or an optical-humanoid-recognition-and-OCR system including the herein-disclosed gateways) captures data. In addition, the system, can monitor, and can include, as part of a score card, surface and environmental conditions such that each leg of a race can have information, including the following data, associated with the leg: localized (to the location of the leg) weather report and/or weather conditions, localized road (or other surface) gradient, highest elevation, lowest elevation, and elevation gradient, and general weather conditions for the entire course.

Using an embodiment of an AI system (automated, partially automated, or manual) such as disclosed herein can allow generating an embodiment of a scorecard such as disclosed herein without the need to retrieve, manually, weather conditions from a historical database, and can provide for the automatic (e.g., partially automatic, fully automatic) generation of score cards for event participants (e.g., runners in a race) at, for example, the end of the event or when participants complete the event (e.g., some runners are faster, and, therefore, finish the race sooner, than other runners).

Said another way, AI “comes into the picture” after the course data is sent to a computer system, such as the cloud, including a database, which can be a permanent large knowledge base of performances that the database stores and characterizes based on demographics such as participant (e.g., runner) weights, heights, or stride lengths. An AI engine can be configured to cross-correlate the data from one race to the large knowledge base. For example, while cloud-based applications can be configured to publish basic statistics (e.g., average race times, variance and standard deviation in race times) for each leg of a running course, and even to correlate runner weights or heights with runner race times and to generate a plot of the same, an AI engine can be configured to generated feedback that is insightful and better than what humans generally code into a software application. And the AI engine can be easy to engage and may not require knowledge of programming languages or SQL (for databases). A prompt tied to an AI engine can be all that is needed to accomplish the “magic” tied to an AI engine. That is, a user (e.g., a participant in a race) can enter, in response to the prompt, only clear instructions in plain English to get started. For example, an AI engine might calculate, or otherwise determine or generate, some analytics for a person (e.g., event participant) and provide feedback stating “if you gain 10 lbs., I see a new distribution curve for your demographics that looks like this and has a mean of 8 min/mile. But I also see a significant drop in the number of active participants in your demographics if you gain 10 lbs.” Consequently, the AI engine can be configured to provide “that little extra” that humans might not even be searching for.

2 FIG. Furthermore, an example of a scorecard generated by an embodiment of an AI system (e.g., an AI engine) disclosed herein, and of information included on the scorecard, is shown in, and described in conjunction with,.

An embodiment of an AI system can perform a more-detailed analysis where cross correlation and impact to performance can be analyzed using AI. For example, an embodiment of the AI system can correlate event data in a multidimensional way and report the AI system's findings.

In an embodiment, an AI system can perform multi-dimensional analytics. For example, if a runner's pace on hills is better than the pace on hills of most other runners but the runner's flat-terrain pace is not better than the flat-terrain pace of most other runners, then the AI system can cross reference runners of similar build and cross reference data to see if weight or height, or any other parameter, is a potential reason why the runner's flat-terrain pace is relatively poor as compared to the flat-terrain base of most other runners. And the AI system also can analyze environmental conditions such as wind, temperature, or humidity to determine the impact of one or more environmental conditions on a runner's performance.

Another embodiment of the AI system can invite users (e.g., runners) to enter a few personal parameters for more detailed analytics on personal traits. Data such as gender, weight, age, and height are optional to enter but if the user does enter such data, the AI system can use the data to create a knowledge base of performance with the traits and other data entered by the user. This knowledge base is published to the users (e.g., event participants such as runners), e.g., when a runner wonders how he might do on a particular course. Or if a runner has participated in a race and wonders how he might do if he drops or loses 10 lbs. of body weight, then the AI system can provide an estimate based on data available to the AI system. For example, an embodiment of the AI system can have access to a large knowledge base of performance vs. demographics, and using data in this knowledge base, can estimate, or otherwise can determine, how a runner's performance can change with minor changes to the runner's age, weight, or other traits or parameters.

1 FIG. 1 FIG. 100 102 104 104 106 106 108 110 112 114 104 108 110 116 1 n 1 n is a mapof an event coursewith gateways-(e.g., Bluetooth® gateways with n=5 in an embodiment) positioned at checkpoints-along the event course and a gatewaywith video equipmentat a start/finish line, according to an embodiment in which the event course is a running-race course (e.g., 1 kilometer (k), 5 k, or 10 k), the gateways can be similar to embodiments of gateways disclosed herein, and the video equipment can be configured to sense when a participant (not visible in) crosses the finish line and to take, automatically, an electronic image or a video of the participant as he/she crosses the finish line; alternatively, a video operatorcan take the image or video. One or more of the gatewaysor, or the video equipment, can be coupled to a cloud-based database or AI engine via a satellite or other internet connection(e.g., wireless broadband or wired (electrical or optical) broadband).

118 102 106 1181 112 106 118 106 106 118 106 106 118 106 106 118 106 106 118 106 106 118 106 106 118 106 112 118 118 118 106 106 102 118 118 1 2 1 2 3 2 3 4 3 4 5 4 3 6 3 2 7 2 1 8 1 3 6 2 3 1 8 Legscan be defined as the distances, along the event course, between consecutive pairs of checkpointsin the direction that the event is conducted (e.g., in the direction that the race is run). For example, a first legcan be defined as the distance from the start/finish lineto the first checkpoint, a second legcan be defined as the distance from the first check pointto the second checkpoint, a third legcan be defined as the distance from the second checkpointto the third checkpoint, a fourth legcan be defined as the distance from the third checkpointto the fourth checkpoint, a fifth legcan be defined as the distance from the fourth checkpointback to the third checkpoint, a sixth legcan be defined as the distance from the third checkpointback to the second checkpoint, a seventh legcan be defined as the distance from the second checkpointback to the first checkpoint, and the eighth and final legcan be defined as the distance from the first checkpointback to the start/finish line. Although certain pairs of the legs(e.g., legsand) define the same section (i.e., between checkpointsand) of the event course, because an event participant runs such a pair of legs in opposite directions and at different times during the event, the same section is counted as two legs for event purposes. Furthermore, the different ones of the legs-can be the same or different lengths.

1 FIG. 1 FIG. 112 108 116 Still referring to, in an embodiment during an event, a participant wearing a beacon (e.g., a Bluetooth®, Wi-Fi®, or RFID beacon according to an embodiment disclosed herein, neither the runner nor the beacon visible in) starts the race at the starting lineand the gatewayrecords the identity of the participant and his/her starting time and can provide the participant identity and starting time to the cloud via the internet connection.

106 106 102 102 102 102 118 118 102 116 1 3 1 3 1 3 1 3 The participant runs toward and past the first, second, and third checkpoints-, and the respective gateways-record the identity of the participant in response to the beacon worn by the participant and record the respective times that the runner arrives at the first, second, and third checkpoints. One or more of the first, second, and third gateways-may determine the respective times it took the participant to traverse the first, second, and third legs-of the course(these leg times may be called “splits” or “split times”), or may provide, via the internet connection, the participant identification and checkpoint times to the cloud for calculation of the first, second, and third leg times.

106 106 118 106 118 106 118 106 118 112 118 102 102 108 106 106 112 102 106 108 118 118 102 116 3 4 4 3 5 2 6 1 7 8 4 1 4 1 4 1 4 8 The participant continues to run from the third check pointto the fourth checkpointvia the fourth leg, from the fourth check point back to the third checkpointvia the fifth leg, from the third checkpoint back to the second checkpointvia the sixth leg, from the second checkpoint back to the first checkpointvia the seventh leg, and from the first checkpoint to the finish linevia the eighth leg. The respective gateways-and the gatewayrecord the identity of the participant in response to the beacon worn by the participant and record the respective times that the runner arrives at the fourth, third, second, and first checkpoints-and the finish line. One or more of the fourth, third, second, and first gateways-or the gatewaymay determine the respective times it took the participant to traverse the fourth, fifth, sixth, seventh, and eighth legs-of the course, or may provide, via the internet connection, the participant identification and checkpoint times to the cloud for calculation of the fourth, fifth, sixth, seventh, and eighth leg times.

110 114 108 Furthermore, the video system, automatically or under control of the operator, can take an image or video of the participant and, for example, offer (e.g., offer to sell) the image or video to the participant via a kiosk or other component that is part of the video system or that is separate from the video system (e.g., part of the gateway).

102 102 108 106 106 118 118 1 4 1 4 1 8 1 FIG. The gateways-andcan provide, for the participant, not only the start time, the times at which the participant arrived at the checkpoints-, the finish times, and the leg-split times, but also can provide other information, such as the local outdoor temperature, weather, time of day, absolute elevation of the checkpoints, respective elevation gradients of the course legs, participant age, participant weight, participant temperature, or participant sex so that an AI engine (not visible in) situated in the cloud and with access to an event database can make calculations (e.g., average course speed, average leg speed) or predictions (e.g., how fast the participant would have run the race had the participant been ten pounds heavier or lighter, had the outdoor temperature or weather been different, had the elevation or elevation gradient been different, or had the time of day been different) according to one or more embodiments described herein.

2 FIG. 2 FIG. 200 202 204 206 208 210 includes a mapof an event coursewith gatewaysdeployed at checkpointsalong the course, an elevational mapof the event course, and, for an event participant (not visible in), a score cardgenerated in response to data captured by the gateways, provided by a cloud database, or both the captured and database data, according to an embodiment.

204 The gatewayscan be gateways according to one or more gateway embodiments disclosed herein.

202 212 204 206 206 204 204 214 214 1 2 12 2 12 1 12 Participants run the event coursein a clockwise direction starting at a start/finish linenear gatewayand passing by checkpoints-and corresponding gateways-and along legs-.

210 214 214 202 1 12 The scorecardincludes, for each leg-, a corresponding time it took the participant to run the leg, the participant's pace during the leg, the length (distance) of the leg, the elevations at the starting point and ending point of the leg and the average slope of the leg, the type of surface(s) (here asphalt) of the event coursealong the leg, and one or more conditions along the leg including whether it was day or night when the runner traversed the leg, the percent cloudiness, the average wind speed and wind direction, and the air temperature.

210 216 218 220 222 The score cardalso can include the participant's total event time, average paceover the entire event, and finishing positionrelative to the total numberof event participants who finished the event.

204 210 208 214 The gatewayscan detect and acquire or capture all of the data that one or more of the gateways, or a cloud computing system, uses to calculate the items on the score card, or at least some of the data (e.g., weather conditions, whether it is day or night, elevational map, or types of surfaces of the legs) can be retrieved from a cloud database.

3 FIG.A 1 FIG. 1 FIG. 3 FIG.A 300 110 112 3021 3022 is an image (e.g., photo)taken by the video system() at the finish line() as one or more (two in) participants-finish the event (e.g., by crossing the finish line), according to an embodiment.

110 304 304 302 302 306 306 304 1 FIG. 1 2 1 2 1 2 A computing system onboard a video system (e.g., the video systemof), or partially or fully located in the cloud, executes a machine-learning model that is trained to detect, and to form bounding boxes-around, the participants-, who are wearing badges, e.g., bibs,-(shown in normal-size and magnified views) on their persons (e.g., clothes such as shirts, shorts, hats, or body parts such as arms, legs) so that the side of each bib with participant-identifying data (e.g., printed or etched thereon) is visible to one or more cameras of the video system. A reason that the computing system can be configured to generate one or more bounding boxes, such as the bounding boxes, around one or more persons (e.g., humanoids, event participants) in an image can be to demonstrate where the computing system has detected one or more persons in an image; once the computing system detects a person in an image, the computing system then can identify the detected person as disclosed herein.

306 306 1 2 Each of the bibs-can be an embodiment of the Bluetooth®, Wi-Fi®, or RFID beacons described herein with participant-identifying information (e.g., a photo, a name, a participant identification number) printed or otherwise present thereon, or each bib can be separate from the Bluetooth®, Wi-Fi®, or RFID beacon worn by the participant.

302 302 110 112 1 2 Using humanoid (e.g., facial) recognition on the partial (e.g., face) or full body of each participant-or on each participant's bib photo, or using another participant—identifying technique such as using optical character recognition (OCR) to derive the participant's identification number from an image of the bib, the video systemcan identify each participant as he crosses the finish line.

110 300 302 306 110 300 3 FIG.A 3 FIG.A The video systemthen can offer for sale, via a kiosk (not visible in) or other suitable device, an electronic or printed version of the photoof the participantcrossing the finish line (the bounding boxes and the magnified views of the bibstypically are not in the photo offered for sale). Or the video systemcan offer an electronic version of the photofor purchase via the identified participant's smart phone or other computing device (neither visible in).

302 300 Each participantthen can decide whether to purchase the photo, and, if he decides to purchase the photo, he can do so via the kiosk, smart phone, or other computing device in a conventional manner.

102 302 110 110 112 110 106 102 1 FIG. 3 FIG.A 1 FIG. 1 FIG. 1 FIG. Furthermore, although, as disclosed herein, one or more gateways() can identify each participantin response to signals emitted by a Bluetooth® or Wi-Fi® beacon (not visible in) worn by the participant, in an embodiment the video systemcan render the gateway() at the finish line unnecessary by using humanoid-recognition/image-recognition/OCR to identify the participants at the finish line() and the times at which the participants each cross the finish line. Or versions or nodes of the video systemcan be installed at each of one or more of the checkpoints() to render unnecessary the corresponding one or more gatewaysand the Bluetooth®, Wi-Fi®, or RFID beacons worn by the participants. Such an embodiment can be used, for example, where event organizers wish to reduce a cost of the event by eliminating the need for the participants to wear Bluetooth®, Wi-Fi®, or RFID beacons, which may be relatively expensive.

354 354 1 2 3 FIG.B In an embodiment, participants wear, on their chests or other body part, a bib with a unique participant identification number such that when they cross the finish line, the video camera scans their bibs' identification numbers and record their crossing times by capturing the moment when runners cross over the finish line. The position of the finish line can be defined in the field of the view of the camera through “threshold crossing” concepts or even through defining x-y coordinates of the start (e.g., one end) of the line to the x-y coordinates of the end (e.g., the other end) of the line that the field(s) of view of one or more of the cameras (e.g., cameras-of) are showing.

For example, a computing system can use a small Nvidia® computer integrated circuit (IC) capable of loading different AI models (Nvidia makes computer ICs capable of varying processing power). For example, one can use a lower-end computer IC that is capable of loading two different AI models into its graphics processor units (GPUs).

304 In an embodiment, a first AI model is configured for participant recognition using humanoid detection. This type of model places a corresponding bounding boxaround each participant within the field of view (FOV) of at least one of the video cameras.

304 306 The second AI model is configured for optical character recognition (OCR), and performs OCR on each of the participants around which the first AI model places a bounding box, for example, by scanning and recognizing the participant identification number on the participant's bib.

112 108 Sometimes events, such as running races, with a smaller number of participants in the trail-running community generally do not have computer systems or other equipment configured to time runners at the starting line. The event video and gateway system provides only a finish time for each participant, and all the participants receive the official start time of the race as their starting time so each participant can determine his race time. But because not every participant crosses the starting lineat the same time at the start of the race, this technique may not be as accurate as a system with gatewaysat the starting/finish line so that the unique starting and finishing times of each participant can be determined. That is, an embodiment of a system disclosed herein can time one or two runners at a time as runners come in and cross over the finish line with gaps between successive groups of one or more runners.

Each participant's finish-line-crossing moment can be captured as an image or video and saved as an image.

304 306 An AI engine can record the finish-line-crossing time of each participant around which the engine generates a bounding boxsuch that every finisher can receive an accurate finish-line-crossing time. If the system fails to optically scan, using OCR, a participant identification number from a bibof a participant, then the system records the participant as an unknown participant, or a participant with an unknown or unobtained participant-identification number.

In an embodiment wherein Bluetooth® tracking technology (e.g., Bluetooth® beacons) is used at the same race, a Bluetooth® signal from the bib/beacon 306 of each participant can provide, to the AI computing system, the participant identification number on the bibs/beacons of the finishers for which the video system did not recognize the participant identification number.

Consequently, in an embodiment, data from the AI computer system can include the unique participant identification number from each of the participant's bibs or beacons and the finish-line-crossing time of each participant.

The AI computer system either can send this data to the cloud where it can be stored, e.g., onto Google® sheets or a website, or can send this data to where it can be ranked to show a multitude of views of the event (e.g., running race) finishers.

Or the AI computer system can be a 100% local system whereby the race director or a human timer supplies a laptop or other computing system for saving information from the AI computer system.

As described herein, such an OCR system can omit RFID tags, Bluetooth® or Wi-Fi® beacons, or any electronic device configured for attachment to an event participant, and still function to identify and time event participants. For example, such an OCR system can be a low-end solution for those who do not want to spend money purchasing one-time-use beacons/tags that are disposed after each event (e.g., a running race).

3 FIG.B 1 3 FIGS.andA 348 112 350 352 352 354 354 108 108 1 2 1 2 1 2 is an isometric viewof the finish lineof, a video systemthat includes multiple (two in this example) video-camera assemblies-each including at least one video camera-, and multiple (two in this example) finish-line gateways-, according to an embodiment.

352 352 350 110 350 352 302 302 112 352 302 354 352 1 2 1 FIG. 3 FIG.B Other than including multiple video-camera assemblies-, the video systemcan be similar to the video systemof. A reason for the video systemincluding multiple video-camera assembliesis that if multiple participants(only one participantvisible in) cross the finish lineat, or approximately at, the same time, one or more participants may partially or fully block another participant from a single video camera's field of view such that the video system cannot take a suitable video or still image of the other participant or of his bib. But with two or more video-camera assemblies, the likelihood of one or more participantspartially or fully blocking another participant from the field of view of all of the cameras is significantly reduced. That is, there is a strong likelihood that each participant is unblocked from, and, therefore, fully exposed within, the field of view of at least one of the video camerasof at least one of the video-camera assemblies.

108 302 302 112 108 108 302 108 3 FIG.B 3 FIG.B Similarly, a reason for including multiple (for example two) gatewaysis that if multiple participants(only one participantvisible in) cross the finish lineat, or approximately at, the same time, one or more participants may partially or fully block a single gatewayfrom signaling another participant's beacon (not visible in) or from receiving the signal emitted by the other participant's beacon such that the single gateway cannot record the other participant's identity, finish time, or other-participant-related data. But with two or more gateways, the likelihood of one or more participantspartially or fully blocking all of the gateways from communicating with the beacon worn by another participant is significantly reduced. That is, there is a strong likelihood that communications are unblocked between the respective beacon of each participant and at least one of the gateways.

108 112 1081 1082 112 1082 Yet another reason for including multiple (for example two) gatewaysis to detect when a participant crosses the finish line. For example, one gatewaycan generate an optical beam over and in alignment with the finish line and another gatewaycan receive the optical beam. When a participant “breaks” the optical beam with his body as he crosses the finish line, the other gatewaydetects this break, can record the time of the break as the time that the participant crosses the finish line, and can provide this finish-line-crossing time to a computing circuit to facilitate the computer circuit identifying the participant and matching the participant to his event-finish time as disclosed herein.

1082 352 112 In an embodiment, the gatewaycan generate, and send to the video-camera assemblies, common or separate signals in response to detecting a participant crossing the finish line, and the video-camera assemblies can capture an image of the participant crossing the finish line in response to the signal(s) such that image is captured as the participant crosses the finish line.

352 112 In another embodiment, the video-camera assembliescan be configured to detect a participant crossing the finish line, to determine the time of the finish-line crossing, and to capture an image of the participant in response to detecting the participant crossing the finish line.

352 352 1 2 In another embodiment, the video-camera assembliesandcan be configured to capture a continuous stream of video images, and to identify one of the images as being an image of a participant crossing the finish line in response to detecting (or to one or more gateways detecting), a time that the participant crosses the finish line, comparing the finish-line-crossing time to the capture times of the video images, and selecting the image of the participant crossing the finish line as the image having a capture time that is closest to the finish-line-crossing time.

4 FIG. 3 FIG.B 348 400 352 108 is the viewofand a computing systemwith which the video-camera assembliesand the gatewayscan communicate in a wired or wireless manner, according to an embodiment.

400 402 404 302 406 408 400 The computing systemincludes an AI processing circuit (e.g., microcontroller, microprocessor, processor), a databaseconfigured to store finishing times (e.g., absolute time and time it took to complete the event course) of the event participants, one or more display devices, and an antenna, internet connection, or other apparatusconfigured to couple the computer system to the internet/cloud. Although shown as being outside of the cloud, one or more of the components of the computer systemcan be implemented in the cloud.

352 108 400 The video-camera assembliesand the start/finish-line gatewaysare configured to capture data as described herein and to provide the captured data to the computing systemin a wired or wireless manner (indicated by the heavy arrows).

402 404 406 404 406 302 For example, the AI processorcan store the participant event-finishing times in the database, which can provide these times to the one or more display devicesfor local display to the event participants and attendees. For example, the databasecan store, and the one or more display devicescan display, not only a corresponding event time for each participant, but other information such as leg/split times, weather conditions, or participant sex, age, identification, or event ranking (e.g., derived from data for multiple similar yet prior events in which the participant participated).

402 302 404 406 The AI processoralso can store the videos and images of the participantsas they cross the finish line 112 in the databaseor in the cloud for sale to, or otherwise for retrieval by, the participants, or can provide these videos or images for display by the one or more display devices.

402 402 302 102 108 110 352 404 1 FIGS. 1 FIGS. Furthermore, the AI processorcan predict “what ifs” as described herein, according to an embodiment. For example, the AI processorcan predict how a participant's event time or leg splits would change if the participant lost 10 lbs. or if the course weather were different. Such predictions for a participantcan be based not only on data captured by the gateways() andand by the video systems() andduring an event, but also can be based on data of the participant's past-event performances stored in the database, in the cloud, or elsewhere.

4 5 FIGS.- 1 FIGS. 402 352 102 108 404 Referring to, the AI processorcan provide or determine information that spans multiple instances of an event (e.g., a 5 k race) based on data captured by the video-camera assembliesand the gateways() andfor a current occurrence of the event and based on data stored in a database (e.g., databaseor a cloud database) regarding past occurrences of the event or of one or more similar events.

5 FIG. 4 FIG. 402 is an embodiment of a depiction of basic analytics on the performances of participants in a running-race event generated by an embodiment of an AI system (for example, a system including the AI processorof) using a knowledge base, according to an embodiment.

500 402 500 402 500 4 FIG. An AI promptis configured to tap into a knowledge base (e.g., a database) of results of one or more occurrences of one or more events and to provide analytics and analysis back to a user (e.g., an event participant), according to an embodiment. For example, the AI processorof, or a virtual AI engine in the cloud, can be configured to receive questions in the AI prompt, to answer the received questions, and provide the answers in the AI prompt. Or one can submit the questions to the AI processoror AI engine in the cloud via other than the AI promptand the AI processor or engine can be configured to generate the AI prompt including both the questions and answers to the questions.

502 500 504 500 106 506 500 20 29 1 FIG. For example, at, the AI promptindicates the fastest male runner for a particular event, and his time, over one or more occurrences of the event. At, the AI promptindicates how many runners crossed a particular checkpoint() of an event course in under one (1) hour over one or more occurrences of the event. And at, the AI promptindicates the slowest female runner in the female age-division over one or more occurrences of the event.

6 FIG. 4 FIG. 1 FIG. 600 400 600 102 is a functional block diagram of an electronic system, such as an electronic computer system, that can be used as the computer systemofor otherwise can be used to perform one or more of the operations or functions described herein, according to an embodiment. One or more components of the systemcan be disposed local to the event course (e.g., the running courseof) or remote from the event course such as in the cloud.

600 602 602 602 402 600 604 602 602 600 606 602 606 600 608 602 608 608 4 FIG. The systemmay include electronic computing circuitry, according to an embodiment. The electronic computing circuitrymay be generally configured to perform various computing functions, which may include, for example, executing specific instructions that may be embodied in software, or performing other specific functions, such as processing data according to the specific instructions, or by other means. For example, the electronic computing circuitrycan execute software instructions that cause the electronic computing circuitry, or other circuitry coupled to the electronic computing circuitry, to function or to operate as the AI processorof. Furthermore, the electronic systemmay also include one or more input devices, which may include an audio input device (e.g., one or more microphones) or a manual input device such as a keyboard, a mouse, a tactile input device, or one or more other similar devices, which may be coupled to the electronic circuitryso that user preferences and instructions may be communicated to the electronic computing circuitry. The electronic systemalso can include one or more output devicescoupled to the electronic circuitry. Suitable output devicesmay include an audio speaker, a display device, as well as other output devices that may depend on a specific function or configuration of the system. One or more data-storage devicesalso can be coupled to the electronic computing circuitryto permit storage and retrieval of data or instructions from storage media, which may be located within the electronic computing circuitry, or located external to the electronic computing circuitry. Examples of suitable storage devicesmay include magnetic storage devices, such as hard-disk devices, or floppy disks, tape cassettes, or other similar devices. Other suitable storage devicesmay include optical storage devices, such as compact disk read-only memory (CDROMs), compact disk read-write (CD-RW) memory devices, digital video disks (DVDs), or solid-state drives or other nonvolatile memory with or without encryption, although other suitable alternatives exist.

Although the foregoing text sets forth a detailed description of numerous different embodiments, it should be understood that the scope of protection is defined by the words of the claims to follow. The detailed description is to be construed as exemplary only and does not describe every possible embodiment because describing every possible embodiment would be impractical, if not impossible. Numerous alternative embodiments could be implemented, using either current technology or technology developed after the filing date of this patent, which would still fall within the scope of the claims.

Thus, many modifications and variations may be made in the techniques and structures described and illustrated herein without departing from the spirit and scope of the present claims. Accordingly, it should be understood that the methods and apparatus described herein are illustrative only and are not limiting upon the scope of the claims.